1. Field of the Invention
The present invention relates to methods and apparatus for measurement of insitu stress in an underground formation traversed by a borehole.
2. Background Information
The need for a tool which could measure the in-situ state of stress in deep wells has increased in recent years. Knowledge of earth stress is required for the planning of stimulation treatments, the prediction of wellbore stability and sand production. Environmental issues, such as the prediction of the long term stability of waste disposal sites, have created new applications for stress measurements at great depth.
Reservoir rocks are commonly sandstones bounded above and below by shale. The difference between the least principal horizontal stress (Sh) in the sandstone and Sh in the shale is dependent on the present tectonic maturity of the basin, the pore pressure, and the mechanical properties of the sandstone and shale. Stress measurements made at closely-spaced intervals in the same borehole indicate that stress magnitudes in sedimentary rocks can vary from bed to bed. Bed-to-bed variation in Sh favors propagation of natural joints in the low stressed beds and acts to prevent joints initiated in the lower stress beds from propagating into beds of higher stress. This phenomenon is exploited by petroleum engineers to contain hydraulic fractures within beds of low stress. A precise knowledge of differences in stress magnitude allows engineers to predict the type of fracture treatment that will assure containment in the reservoir beds. However, precise stress magnitude data are rarely obtained in shales. Instead it is commonly assumed that the least principal horizontal total stress in shales is greater than in adjacent reservoir rocks.
Various techniques have been proposed to measure in-situ stress. Perhaps the most reliable to date for measuring stresses at great depth is the micro-hydraulic fracturing technique. This technique uses the pressure response obtained during the initiation, the propagation and the closure of a hydraulic fracture to determine the state of stress.
In this technique, an interval is isolated using a packer arrangement. A fluid is injected in the interval at a constant flow rate until the wellbore is pressurized sufficiently to initiate a tensile fracture. The fracture initiates and propagates in a direction normal to the minimum stress. Fracture initiation is often recognized by a breakdown on a pressure vs. time record at a pressure termed the "breakdown" pressure, though fracture initiation may occur before the pressure breakdown.
Injection continues after the initial breakdown until the pressure stabilizes. Injection is then stopped and the pressure allowed to decay. The fracturing fluid is often a low viscosity fluid, such as mud or water. A quantity of fluid dependent upon the formation interval size (e.g., usually less than 400 liters) is injected into the formation at flow rates ranging from 1 liter/minute to 100 liters/minute. Several injection/fall-off cycles are usually performed until repeatable results are obtained. A down-hole shut-off tool is sometimes used to shut in the well and minimize any wellbore storage effect. Careful monitoring of the shut-in behavior is required to determine the minimum stress.
The instantaneous shut-in pressure has often been assumed to approximate the minimum stress, though errors of the order of several MPa may result. In permeable formations, where the fracturing fluid leaks off from the fracture face, the minimum stress is better measured by the point at which the pressure decline deviates from a linear dependence on the square root of shut-in time. This technique could also be in error and alternate methods have been developed to estimate the minimum stress, such as the step rate test and the flow back test.
In a step rate test, the injection is increased by steps until the pressure response indicates that a fracture is widely open. Analysis of the propagating pressure vs. flow rates leads to an estimation of minimum stress. The flow back test consists of pumping the fluid out of the fracture once the injection has been stopped. Closure is determined from a change of the pressure response behavior. The closure stress is taken as a measure of the minimum stress.
Attempts have also been made to determine the intermediate stress (often the maximum horizontal stress) from the breakdown pressure. The breakdown is due to the tensile strength of the rock and the stress concentration induced around the well bore. The breakdown pressure Pb is predicted using linear isotropic elasticity and assuming a non-penetrating fluid by the Hubbert and Willis breakdown equation: EQU Pb=3Sh-SH+T-P.sub.p
where SH and Sh are the maximum and minimum horizontal principal stresses, respectively, T is the tensile strength and P.sub.p is the pore pressure of the rock. For injection cycles which follow the first injection cycle, the breakdown corresponds to the reopening of the fracture, and T is then effectively equal to zero. As Sh have been determined from the closure, this equation can be used to estimate the intermediate stress. However this estimation is often poor: the fluid penetrates the fracture before the fracture re-opens, the assumption of linear isotropic elasticity does not apply, the wellbore is not aligned with a principal stress direction or the re-opening pressure is obscured by viscous effects.
A better approach to estimate the complete state of stress is to re-open a pre-existing fracture or a discontinuity. With this method, the closure stress is determined on pre-existing fractures by performing a series of step rates and shutins. The fluid is injected at a very low flow rate (e.g., less than 0.5 liter/minute) to percolate the pre-existing fracture. A clear breakdown is rarely observed, because the injection fluid penetrates the fracture before the opening occurs. The closure stress is a measure of the stress normal to the fracture plane. Measurements made on fracture planes with various dips and strikes allow the complete state of stress to be determined.
A drawback of the open hole hydraulic fracturing technique is that communication between the test interval and the borehole annulus above/below the test interval is often observed during the pressurization phase, preventing the test being carried out properly. Because of the communication problem, cased hole stress tests are often carried out. Cased hole tests are also preferred for operational and safety reasons. Except for the need to perforate the casing (usually a 2 foot interval is perforated), the technique is similar to the open hole hydraulic fracturing technique. Stress measurements in cased holes have disadvantages relative to open hole measurements: fracture orientation and width are hidden by the casing, the fracture may propagate in the cement, breakdown pressures are often much higher than those obtained in open hole, breakdown pressure cannot be easily interpreted (the Hubbert equation does not apply due to the existence of the casing and perforation) and, especially in a petroleum environment, operators are unwilling to perforate the casing in non-productive layers.
Another approach to measuring in-situ stress employs an instrumented, inflatable packer to initiate fractures in the rock without injection of fluid in the rock. U.S. Pat. No. 4,733,567 to Serata; O. STEPHANSSON, Sleeve Fracturing for Rock Stress Measurement in Boreholes, SYMPOSIUM INTERNATIONAL IN SITU TESTING, Volume 2, 571-578, Paris, 1983. While the packer-fracturing technique as proposed thus far has advantages over the hydraulic fracturing technique, its utility is limited by the lack of means for determining fracture orientation and other features needed to obtain useful measurements deep in the earth.